Elvee: In post 106 you state that ...Q7 has to be biased... The way Q7 is drawn it is a diode. Could you elaborate please.
Thanks, E
Thanks, E
It is indeed connected as a diode, and for the servo to operate and remain in its linear range, that diode has to be forward-biased by the collector current from Q13.
That diode can then play its role of compensating Q2's Vbe.
Elvee, I am going to put together a square wave generator hopefully by the end of the week. I have a lot going on, so don't be surprised if it is a bit longer.
I am also in the process of putting together a +-35VDC power supply for these and altering the resistor value you spoke of. When that happens I will let you know how it goes. Still playing this every day, and still sounds simply amazing.
I am also in the process of putting together a +-35VDC power supply for these and altering the resistor value you spoke of. When that happens I will let you know how it goes. Still playing this every day, and still sounds simply amazing.
This is not essential, but it will allow to you to fine-tune the compensations, if required
I am glad you enjoy it, that's the way it's supposed to work.
Hi odysseybmx414, I'm following your review this amplifier with - +35vdc, what the updates now and how it sound? I'm waiting for your next review. Thanks.
Regards,
Boyet
Actually, I have a PCB sitting in etchant right now. It is 7 2200uF capacitors per rail. Just waiting on a mouser order which should arrive any day. I will review in a few days.
Elvee, forgive my ignorance, I tried to figure it out for myself. At +-35V, will I have to change the BC546/556Cs that I am using?
I am and have been using MJW21196. I couldn't speak of this circuit using 2n3055.
I wasn't sure of the actual voltages the small signal transistors are seeing.
Of the BC's, only one will see the full supply voltage: Q13.
At a total supply of 70V, it will see ~67V, and the BC546's Vceo is rated at 65V, thus in theory too much.
But if +/-35V really is the upper limit of your supply, you run no risk using one.
If on the other hand there is a risk your supply drifts higher than 70V, you would be well advised to use a 100V or 80V type, like the MPSL01 or the 2N5550 for example.
Note that Q12 should preferably be identical to Q13.
Other BC's see at most half the total supply and need not being substituted.
The BD's and the 2N3019 are OK too.
Alright, thanks. Also note that I do not use BD's. I'm using 2SB649's as the drivers to the MJW21196's.
So, the changes will be:
--Power supply to +-35VDC
--Q12,Q13 to 2N5551 (Or is KSC1845E better? (much higher hFE, but low Ic))
--R21 to 56K
--D8,D9 changed from 2x12V to 2x15V
Should be up and running within two or three days hopefully.
OK, no problem regarding the 2SB649
Those transistors are just auxiliary devices: they aren't directly in the signal path, they just ensure it's in a good condition and they have little influence on the quality.
Small signal devices are more than enough, and if they have a high Hfe, so much the better, so I would opt for the KSC's
Good.
Elvee. About the power supply. I finished it today. The transformer is supposed to be 25.5 0 25.5. Maybe because I am using low voltage drop diodes, but I am getting higher supply voltage than anticipated.
Plugging it in unloaded gave me a total of 82VDC (+-41V). I then found a high wattage 220R resistor to draw ~ 360mA (depending on sag) from the power supply. This dropped the total voltage to 79V or +- 39.5V.
I suppose I will use 2 18V 1W devices for the zener diodes, but I am assuming I will need a slightly different resistor value for R21 now.
Also, I am assuming I can still keep BC546/556 for all the small signal transistors except for Q12,13 right? I got my mouser order and will use KSC1845F for those. (120V Vceo).
Am I right to understand that the MJW21196's still have plenty of SOA left to spend? I am confident this heatsink is also adequate. The amp won't be pushed very hard.
Thanks again. Looking forward to hearing music from this amp once again.
EDIT: the transformer must have been mislabeled.
Plugging it in unloaded gave me a total of 82VDC (+-41V). I then found a high wattage 220R resistor to draw ~ 360mA (depending on sag) from the power supply. This dropped the total voltage to 79V or +- 39.5V.
I suppose I will use 2 18V 1W devices for the zener diodes, but I am assuming I will need a slightly different resistor value for R21 now.
Also, I am assuming I can still keep BC546/556 for all the small signal transistors except for Q12,13 right? I got my mouser order and will use KSC1845F for those. (120V Vceo).
Am I right to understand that the MJW21196's still have plenty of SOA left to spend? I am confident this heatsink is also adequate. The amp won't be pushed very hard.
Thanks again. Looking forward to hearing music from this amp once again.
EDIT: the transformer must have been mislabeled.
Last edited:
Better diodes could only account for max. ~0.5V or so.
400mW zener diodes are sufficient.
For R21, anything between 47K and 68K will do
No problem there
You are right, your MJW's could handle a 4R load under those conditions.
They are real beasts.
The heatsinks are ample for normal musical signals, even at high volume
I don't think so: transformers have their output voltages specified at their full capacity, your mains voltage could be a bit on the high side, and your transformer might have been somewhat overdesigned too. 🙂
Not bad, you'll get some extra power for free.
OK, The amp has been running for over an hour now.
The heatsink gets comfortably warm at low listening levels but nothing more.
I guessed on 68k when the anticipation got the best of me and I had to get it hooked up. Good guess.
The subjective sound impression:
I didn't think it could possibly sound better, even with a properly filtered power supply with ample capacitance. It already pleased me more than anything I had listened to. Layers of detail, crystal clear voices. I can now make out what a singer is saying easily on songs where I once thought it was impossible without reading the lyrics. There is no 'fog' around individual notes. You can clearly make out the 'silence' that is a vital part of any sound. More important to some, listening to just about anything creates a very good feeling. Instantly brightens your mood. Even songs you may not particularly like.
So, at ~+-39V, with the transistors and circuit configuration (comp. caps) I am using, I couldn't possibly be happier.
I don't have square wave scope measurements, and honestly it may be a while. I am working on a move. But, looking at the output with a scope there is no oscillation, and I have used several different speakers and cable lengths, volume levels.
The parts count is small, and most people have these parts laying around. I would build this amp if I were you.
odysseybmx414, Thank you for very nice reviewed of " ♫♪ My little cheap Circlophone© ♫♪ " designed by Elvee..I will build this one. Thanks again..
Boyet
It's good to hear such appraisal, it confirms my own opinions in every respect (but of course, I don't have an objective view on the matter 😀):
A very healthy, hard working little amplifier, as easy as a chip amp but capable of near class A quality.
I have also re-measured the THD, but this time with proper layout, grounding, etc, and found it to be less than half the figure I initially announced: well under 0.002%. OK, not top class, but respectable anyway.
Note: when increasing the supply voltage to a circuit, make sure you also increase the rating of the capacitors!
Hah. I heard a bang and thought the worst. I'm surprised that capacitor lasted as long as it did, running at 5V above its rating for about 12 hours.
Hah. I heard a bang and thought the worst. I'm surprised that capacitor lasted as long as it did, running at 5V above its rating for about 12 hours.
No, but Wakibaki has created a PCB, the files can be found here:
http://www.diyaudio.com/forums/solid-state/189599-my-little-cheap-circlophone-6.html#post2590488
You can view/print them using GC-prevue f.e. (free):
GraphiCode - The Original Gerber viewer
There are certainly other programs/applications to convert the files into a pdf, but I don't have any example, maybe another member could help you?
Last edited:
C3 (DC servo cap) is the main one that worries me.
And I got this exact same problem in most of own simulated designs.
Sim won't complain when DC offset reverse biases this polarized cap.
Even if it leaks, might never know, just won't roll off the DC anymore.
I suggest a series of two 1000uF polarized caps, back to back.
With the midpoint pulled down to the negative rail though 47K.
Bypass this stack of polarized elko's with a good 1uF film cap....
Last edited: